Method and device for determining defined value of hypotension in hemodialysis and medium thereof

Abstract

The invention relates to the technical field of hemodialysis monitoring, in particular to a method for determining a defined value of hypotension in hemodialysis, which comprises the following steps: constructing a data set; counting and analyzing the data set based on the definition of dialysis hypotension and a data analysis strategy, obtaining an associated index factor in combination with expert opinions, and inputting corresponding data into an AI prediction model for analysis to obtain an AUC value and an accuracy rate; determining a set of dialysis hypotension defined values based on the AUC values and the accuracy; verifying the set and a complication report recorded by the medical staff to obtain a subset of the set, and judging whether a dialysis hypotension event set in the data set completely contains a dialysis hypotension event recorded by the medical staff or not when each definition value in the current subset is used as a dialysis hypotension standard of a verification set data record; and if so, taking the maximum value in the subset of the dialysis hypotension defined value set as the dialysis hypotension standard defined value. According to the method, the defined value of dialysis hypotension suitable for the current patient group can be obtained, and the prediction accuracy is high.

Description

Technical Field

[0001] The present invention relates to the technical field of hemodialysis monitoring, and particularly to a method, device and medium for determining the definition value of hypotension during hemodialysis. Background Art

[0002] Predicting the probability of a patient developing hypotension during dialysis requires a definition of hypotension during dialysis. However, there is currently no unified international consensus on the definition of in - dialysis hypotension (IDH). This means that different definitions of in - dialysis hypotension have a crucial impact on the establishment of prediction models for the probability of in - dialysis hypotension, clinical analysis during the dialysis process, and clinical evaluation of prediction results. Therefore, there is an urgent need to develop a method for determining the definition value of hypotension during hemodialysis to explore the definition of dialysis hypotension in the patient population and improve the accuracy of prediction results. Summary of the Invention

[0003] The technical problem to be solved by the present invention is: to solve the technical problem that there is no unified consensus on the definition of in - dialysis hypotension in the prior art and the prediction accuracy of hypotension is low. The present invention provides a method for determining the definition value of hypotension during hemodialysis, which can obtain the definition value of dialysis hypotension suitable for the current patient population and improve the accuracy of predicting the probability of in - dialysis hypotension.

[0004] The technical solution adopted by the present invention to solve its technical problems is: a method for determining the definition value of hypotension during hemodialysis, the method comprising the following steps:

[0005] S1, obtaining sample case data and pre - processing the sample case data to construct a data set;

[0006] S2, statistically analyzing the data set, and based on the definition of dialysis hypotension and the data analysis strategy, combining expert opinions to obtain the associated index factors of the hypotension definition;

[0007] S3, inputting the data corresponding to the associated index factors of the hypotension definition into an AI prediction model for processing and analysis to obtain the AUC value and accuracy rate;

[0008] S4, determining a set of dialysis hypotension definition values based on the AUC value and accuracy rate;

[0009] S5, verifying the set of dialysis hypotension definition values with the report on dialysis hypotension complications recorded by medical staff to obtain a subset of the set of dialysis hypotension definition values, such that when each definition value in the current subset is used as the dialysis hypotension standard for verifying the data record, it is determined whether the set of dialysis hypotension events in the data set completely includes the in - dialysis hypotension events recorded by the medical staff;

[0010] If it is included, the maximum value in the subset of the dialysis hypotension definition value set is used as the standard definition value of dialysis hypotension;

[0011] If it is not included, find a maximum dialysis hypotension definition value from the set of dialysis hypotension definition values such that all values in the set of blood pressure reduction values of the hypotension record in the complication report are greater than this dialysis hypotension definition value. This dialysis hypotension definition value is the standard definition value of hypotension for the current pre-dialysis systolic blood pressure interval.

[0012] According to an embodiment of the present invention, the step S1 includes: collecting the record data before and during dialysis from the LIS system, HIS system, and dialysis management system to form the sample case data, and the sample case data includes patient dialysis basic data and laboratory data;

[0013] The preprocessing of the sample case data includes: collecting and cleaning the patient dialysis basic data and laboratory data based on time logic to form a data set within an effective time period.

[0014] According to an embodiment of the present invention, the step S2 includes the following steps:

[0015] S21, set the blood pressure drop ratio value α, and perform positive and negative case classification of hypotension on the data in the data set according to the dialysis hypotension definition corresponding to the blood pressure drop ratio value α. After classification, hypotensive dialysis data and non-hypotensive dialysis data are formed;

[0016] S22, perform data optimization on the hypotensive dialysis data and non-hypotensive dialysis data;

[0017] S23, through statistical analysis of the distribution laws of the optimized hypotensive dialysis data and non-hypotensive dialysis data, combined with expert opinions, obtain the characteristic index factor variables related to whether dialysis hypotension occurs;

[0018] S24, screen the characteristic index factor variables to obtain the associated index factors of the hypotension definition.

[0019] According to an embodiment of the present invention, the step S3 specifically includes the following steps:

[0020] S31, obtain the optimal blood pressure drop ratio value α from the dialysis systolic blood pressure value interval [α imin , α imax ; ir ;

[0021] S32, input the data corresponding to the associated index factors of the hypotension definition into the AI prediction model;

[0022] S33, the AI prediction model is based on the optimal blood pressure drop ratio value α irRe - fit and verify the data corresponding to the associated index factors of the hypotension definition using data analysis strategies to obtain the AUC value and accuracy rate.

[0023] According to an embodiment of the present invention, step S4 specifically includes the following steps:

[0024] If the AUC value is less than or equal to the preset value, adjust the dialysis hypotension definition value and return to step S2;

[0025] If the AUC value is greater than the preset value, save the current dialysis hypotension definition value, update the data set or optimize the AI prediction model, and repeat steps S1 to S4 until the AUC value is greater than or equal to the expected value and the blood pressure drop ratio value α reaches the strictest standard definition value. Then, stop optimizing the data set or the AI prediction model, obtain the AUC values and accuracy rates corresponding to all saved dialysis hypotension definition values, compare them with the preset conditions to obtain a comparison result, and form a set of dialysis hypotension definition values based on the comparison result.

[0026] According to an embodiment of the present invention, in step S31, α imin The calculation formula is:

[0027] α imin = 20 / ξ imax (ξ imax > 110 mmHg);

[0028] α imin = 10 / ξ imax (ξ imax <= 110 mmHg);

[0029] α imax The calculation formula is:

[0030] α imax = (ξ imax - 90) / ξ imax ;

[0031] Where ξ imax is the maximum value of the current dialysis systolic blood pressure value range, α imin is the initial drop ratio value for each interval, and α imax is the maximum drop ratio value for each interval.

[0032] According to an embodiment of the present invention, the construction of the AI prediction model includes the following steps:

[0033] Build a data model and update the hypotension label value in the data record set according to the current definition of dialysis hypotension;

[0034] Use an algorithm model to fit the data corresponding to the associated index factors defined for hypotension, and use the particle swarm optimization algorithm to tune the parameters of each algorithm model;

[0035] Compare the prediction results of multiple locally optimal algorithm models, and select the locally optimal algorithm model as the mining model;

[0036] Input data into the mining model for inference, check and calculate and screen the inference results and the records of hypotensive complications in the dialysis record report to obtain the optimal inference result and the optimal dialysis systolic blood pressure drop ratio value, and verify the mining model based on the inference result. If the inference result is satisfied, the mining model is an AI prediction model.

[0037] According to an embodiment of the present invention, in step S22, principal component analysis is used to optimize the data of hypotensive dialysis data and non-hypotensive dialysis data by repeated dimensionality reduction and dimensionality increase operations.

[0038] A computer device, comprising:

[0039] A processor;

[0040] A memory for storing executable instructions;

[0041] Wherein, the processor is configured to read the executable instructions from the memory and execute the executable instructions to implement the method for determining the hypotension definition value in hemodialysis as described above.

[0042] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor is caused to implement the method for determining the hypotension definition value in hemodialysis as described above.

[0043] The beneficial effect of the present invention is that a method for determining the hypotension definition value in hemodialysis according to the present invention can obtain a definition value of dialysis hypotension suitable for the current patient group, solve the fragmentation of the discrimination of the same dialysis record by multiple definitions, and the new definition and analysis strategy can make the discrimination of dialysis hypotension cases have standard unity, improve the accuracy of predicting the occurrence probability of hypotension during dialysis, provide a basic reference for studying the causes of hypotension and formulating corresponding preventive measures, help prevent the occurrence of dialysis hypotension clinically, and have very important significance for improving the dialysis quality of hemodialysis patients and ensuring the safety of dialysis patients during the dialysis process. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The present invention will be further described below with reference to the drawings and embodiments.

[0045] Figure 1 It is a schematic flowchart of the method according to Embodiment 1 of the present invention.

[0046] Figure 2 It is a data density analysis diagram of all index factors in the data set of the first embodiment of the present invention.

[0047] Figure 3 It is a schematic structural diagram of the computer device according to the second embodiment of the present invention.

[0048] In the figure, 10 is the computer device; 1002 is the processor; 1004 is the memory; 1006 is the transmission device. Detailed implementation manners

[0049] Now, the present invention will be further described in detail with reference to the accompanying drawings. These drawings are all simplified schematic diagrams, only showing the basic structure of the present invention in a schematic manner, so they only show the components related to the present invention.

[0050] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, so it should not be construed as a limitation of the present invention. In addition, the features defined as "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, unless otherwise specified, the meaning of "plurality" is two or more.

[0051] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected, or indirectly connected through an intermediate medium, and it may be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0052] Embodiment 1

[0053] The embodiment of the present application provides a method for determining the hypotensive definition value in hemodialysis, as Figure 1 shown, the method includes the following steps:

[0054] S1, obtain sample case data and preprocess the sample case data to construct a data set;

[0055] S2. Conduct statistical analysis on the dataset. Based on the definition of dialysis hypotension and the data analysis strategy, combined with expert opinions, obtain the associated index factors for the definition of hypotension.

[0056] S3. Input the data corresponding to the associated index factors for the definition of hypotension into the AI prediction model for processing and analysis to obtain the AUC value and accuracy rate.

[0057] S4. Determine the set of dialysis hypotension definition values based on the AUC value and accuracy rate.

[0058] S5. Verify the set of dialysis hypotension definition values with the reports on dialysis hypotension complications recorded by medical staff to obtain a subset of the set of dialysis hypotension definition values. When each definition value in the current subset is used as the dialysis hypotension standard for the data records in the validation set, determine whether the set of dialysis hypotension events in the dataset completely includes the in - dialysis hypotension events recorded by medical staff.

[0059] If it includes, use the maximum value in the subset of the set of dialysis hypotension definition values as the standard definition value for dialysis hypotension.

[0060] If it does not include, find a maximum dialysis hypotension definition value from the set of dialysis hypotension definition values such that all values in the set of blood pressure reduction values for hypotension records in the complication report are greater than this dialysis hypotension definition value. This dialysis hypotension definition value is the standard definition value for hypotension in the current pre - dialysis systolic blood pressure range.

[0061] In the embodiment, step S1 includes: Collect the pre - dialysis and in - dialysis record data from the LIS system, HIS system, and dialysis management system to form sample case data. The sample case data includes patient dialysis basic data and laboratory data; the patient dialysis basic data includes: age, weight, pre - dialysis and in - dialysis systolic blood pressure records, pre - dialysis and in - dialysis diastolic blood pressure records, ultrafiltration, heart rate, diagnosis, and complication records; the laboratory data includes blood routine data.

[0062] The pre - processing of the sample case data includes: Collect and clean the patient dialysis basic data and laboratory data based on time logic to form a dataset within an effective time period to ensure the accuracy and integrity of the data and avoid interfering with the AI prediction model.

[0063] Collect and clean the patient dialysis basic data and laboratory data based on time logic, that is, collect and clean the patient dialysis basic data and laboratory data according to the patient ID and dialysis time. For example, collect and clean the dialysis basic data and laboratory data within the one - month period before the dialysis time of the patient based on the patient ID.

[0064] In the embodiment, step S2 includes the following steps:

[0065] S21. Set the blood pressure drop ratio value α, and classify the positive and negative cases of hypotension for the data in the dataset according to the definition of dialysis hypotension corresponding to the blood pressure drop ratio value α. After classification, dialysis data with hypotension and dialysis data without hypotension are formed.

[0066] S22. Optimize the dialysis data with hypotension and the dialysis data without hypotension to improve the rationality and accuracy of data distribution. For example, use principal component analysis (PCA) to perform repeated dimensionality reduction and dimensionality increase operations on the dialysis data with hypotension and the dialysis data without hypotension to achieve the purpose of filtering and noise reduction.

[0067] S23. Obtain the characteristic index factor variables related to whether dialysis hypotension occurs by statistically analyzing the distribution laws of the optimized dialysis data with hypotension and the dialysis data without hypotension, and combining expert opinions.

[0068] S24. Screen the characteristic index factor variables to obtain the associated index factors for the definition of hypotension.

[0069] Further, step S23 specifically includes: applying the SPSS 22.0 statistical software and the statsmodels statistical analysis library of the Python platform for statistical analysis. Measurement data are expressed as mean ± standard deviation, and t-tests and analysis of variance are used for between-group comparisons. Measurement data are expressed as percentages, and chi-square tests are used for between-group comparisons. For the significance test of the categorical data characteristic items and the quantitative target (ultrafiltration volume), the T-test is used, and for the significance test of the quantitative data characteristic items and the quantitative target (ultrafiltration volume), the variance test is used. Determine the optimal characteristic parameter set through multivariate statistical analysis (the simultaneous influence of multiple factors on dialysis hypotension) and multi-factor interaction analysis.

[0070] Through the screening of multivariate analysis based on random forest and XGBoost, it is determined that age, pre-dialysis weight, dry weight, pre-dialysis systolic blood pressure, pre-dialysis diastolic blood pressure, pre-dialysis pulse, target dehydration, monocytes, eosinophils, pre-dialysis calcium, hematocrit, platelet count, pre-dialysis serum creatinine, pre-dialysis urea nitrogen, fasting blood glucose, and pre-dialysis uric acid are the characteristic index factor variables related to dialysis hypotension; for example, there are 6742 cases of dialysis record data (dialysis report plus laboratory report data) for each factor's corresponding dataset. After processing, 63536001 cases are determined as non-hypotension record data, and 745741 cases are determined as hypotension record data. The data density analysis of all index factors in the dataset is as Figure 2 shown.

[0071] Verify whether the characteristic index factor variables of the factors screened from the above dataset are statistically significant. Retrospectively analyze the data of the patients in the dataset. 70,986,742 cases of data will be included and randomly and evenly sampled into a model training set and a model validation set. In the model training set, first, use the univariate statistical analysis method to screen out multiple factor indicators affecting dialysis hypotension, and take the indicator items with statistical significance of P < 0.05 as the variable factors of the model; at the same time, conduct a multivariate statistical analysis on the key defined values related to hypotension, that is, the blood pressure drop value (or the lowest blood pressure value during dialysis can be used), and take the indicator factors with clinical significance of P < 0.2 as the model influencing factors. The factor indicators and analyses of the two types of statistics are shown in Table 1.

[0072] Table 1

[0073] Index factor P value of single-factor T test P value of multi-factor analysis Age 0.001 0.1 Pre-dialysis weight (Kg) 0.973 0.001 Dry weight (Kg) 0.597 0.06 Pre-dialysis systolic blood pressure (mmHg) 0.536 0 Pre-dialysis diastolic blood pressure (mmHg) <0.05 0 Heart rate (beats / min) 0.0001 0 Ultrafiltration volume (Kg) <0.05 0 Monocyte count (×10^9 / L) 0.445 0.2 Eosinophil count (×10^9 / L) 0.864 0.4 Serum calcium (mmol / L) 0.965 0.05 Hematocrit (%) <0.05 0 Platelet count (×10^9 / L) 0.121 0.2 Serum creatinine (umol / L) 0.461 0.0002 Urea nitrogen (mmol / L) 0.001 0.0000006 Glucose (mmol / L) <0.05 0.00007 Uric acid (umol / L) 0.02 0.000003

[0074] Among them, for the test indicator items with P < 0.05 in the univariate analysis, the indicator differences have statistical significance. In the multivariate analysis, the test indicator items with P < 0.2 are considered to have clinical significance by the expert group. The eosinophil did not meet the statistical requirements in both univariate and multivariate analyses. After the subsequent data indicators are stable, it needs to be removed from the set of associated characteristic index factor variables.

[0075] In the embodiment, step S3 specifically includes the following steps:

[0076] S31, obtain the optimal blood pressure drop ratio value α imin , α imax from the dialysis systolic blood pressure value range [α ir ;

[0077] S32, input the data corresponding to the associated indicator factors of the hypotension definition into the AI prediction model;

[0078] S33, the AI prediction model re - fits and validates the data corresponding to the associated indicator factors of the hypotension definition based on the optimal blood pressure drop ratio value α ir and the data analysis strategy to obtain the AUC value and the accuracy rate.

[0079] In the embodiment, step S4 specifically includes the following steps:

[0080] If the AUC value is less than or equal to the preset value, adjust the dialysis hypotension definition value and return to step S2;

[0081] If the AUC value is greater than the preset value, save the current definition value of dialysis hypotension, update the data set or optimize the AI prediction model, and repeat steps S1 to S4 until the AUC value is greater than or equal to the expected value and the blood pressure drop ratio value α reaches the strictest standard definition value (the maximum value of the blood pressure drop ratio), then stop optimizing the data set or the AI prediction model, obtain the AUC values and accuracies corresponding to all the saved dialysis hypotension definition values, compare them with the preset conditions to obtain the comparison results, and form a set of dialysis hypotension definition values based on the comparison results.

[0082] In the embodiment, the construction of the AI prediction model includes the following steps:

[0083] Construct a data model and update the hypotension label values in the data record set according to the current definition of dialysis hypotension;

[0084] Use the algorithm model to fit the data corresponding to the associated index factors of the hypotension definition, and use the particle swarm optimization algorithm to optimize the parameters of each algorithm model;

[0085] Compare the prediction results of multiple local optimal algorithm models, and select the local optimal algorithm model as the mining model;

[0086] Input data into the mining model for reasoning, check and calculate and screen the reasoning results and the hypotension complication records in the dialysis record report to obtain the optimal reasoning result and the optimal dialysis systolic blood pressure drop ratio value, verify the mining model based on the reasoning results, and if the reasoning results are satisfied, the mining model is the AI prediction model.

[0087] In the embodiment, in step S22, principal component analysis is used to optimize the hypotensive dialysis data and non-hypotensive dialysis data by repeatedly reducing and increasing the dimensions of the data.

[0088] In this embodiment, different values of the blood pressure drop during dialysis ΔSBP define different recognition results of hypotension relative to different ranges of pre-dialysis systolic blood pressure values, either too loose or too strict. In the definition of dialysis hypotension:

[0089] When ΔSBP is 20 mmHg, ΔSBP is the minimum value, and at this time, it can be called the loosest definition of dialysis hypotension;

[0090] When ΔSBP is 90 mmHg, ΔSBP is the minimum value, and at this time, it can be called the strictest definition of dialysis hypotension.

[0091] It can be seen that, based on the strictest definition standard, i.e., when ΔSBP is 90 mmHg, the vast majority (more than 98%) of dialysis records will be classified as non-low-normal pressure category. On the contrary, based on the loosest definition standard, i.e., when ΔSBP is 20 mmHg, nearly half of the dialysis records will be classified as low-normal pressure category. Using either value at both ends of the ΔSBP value range as the definition standard value for hypotension does not conform to the medical common sense that the incidence of dialysis hypotension is 10% - 30%. In the range where the pre-dialysis systolic blood pressure is greater than 140 mmHg, judged by the loosest definition of dialysis hypotension, it is possible that the vast majority of non-intradialysis hypotension events (negative cases) will be recorded as intradialysis hypotension events (positive cases). Conversely, judged by the strictest definition of dialysis hypotension, it is possible that the vast majority of intradialysis hypotension events (positive cases) will be recorded as non-intradialysis hypotension events (negative cases). In the range where the pre-dialysis systolic blood pressure is less than 110 mmHg, judged by the loosest definition of dialysis hypotension, it is possible that the vast majority of intradialysis hypotension events (positive cases) will be recorded as non-intradialysis hypotension events (negative cases). Because when comparing with the hypotension events during dialysis recorded by medical staff, in the current situation, the ΔSBP value is too strict. When a hypotension event occurs within the current range and the corresponding ΔSBP value is almost less than 20 mmHg, judged by the strictest definition of dialysis hypotension, it is possible that some non-intradialysis hypotension events (negative cases) will be recorded as intradialysis hypotension events (positive cases). Just like in the range where the pre-dialysis systolic blood pressure is greater than 90 mmHg and less than 100 mmHg, when the blood pressure during dialysis drops to 90 mmHg, the patient will not have hypotensive complications.

[0092] Then it can be known that different pre-dialysis systolic blood pressure ranges should correspond to an unknown ΔSBP value to define dialysis hypotension. Therefore, within a certain pre-dialysis systolic blood pressure value range ξa - ξb, a corresponding value of ΔSBPi can be found, so that the hypotension record data within the range conforms to the current definition. It can be understood that the difference between the pre-dialysis systolic blood pressure and the minimum systolic blood pressure during dialysis in the hypotension records within the range is greater than or equal to this definition-related value ΔSBPi.

[0093] Among the dialysis records (hypotension complication reports) recorded by medical staff as dialysis hypotension, as the pre-dialysis systolic blood pressure value increases, the ratio of the corresponding ΔSBP to the pre-dialysis systolic blood pressure value, that is, the blood pressure drop ratio value α, will also increase. Using the clustering analysis and mining method, all dialysis record data sets are classified according to the pre-dialysis systolic blood pressure range. The result is that the data set is divided into 6 categories, that is, six ranges, and the pre-dialysis systolic blood pressure ξ corresponding to each range is the average value or boundary value of the current range. According to the previous analysis, the pre-dialysis systolic blood pressure analysis range less than 90 mmHg is not included in the analysis scope. The initial value assignment method for the blood pressure drop ratio value α is the ratio of the loosest standard (pressure difference: 20 mmHg) to the maximum value of the pre-dialysis systolic blood pressure value range, that is, the calculation formula is as follows:

[0094] α imin= 20 / ξ imax (ξ imax > 110 mmHg)

[0095] where ξ imax is the maximum value of the current dialysis systolic blood pressure value range, and α imin is the initial decrease ratio value for each range.

[0096] It should be noted that, according to the analysis and comparison with the recorded data set of dialysis hypotensive complication events, for the pre-dialysis systolic blood pressure value range less than or equal to 110 mmHg, the strictest standard is that the pre-dialysis systolic blood pressure drops to 80 mmHg (statistically based on the current defined standard, the lowest systolic blood pressure during dialysis in the dialysis hypotensive complication events is all greater than 80 mmHg), the loosest standard for the 90 - 100 mmHg range is that the pre-dialysis systolic blood pressure drops to 90 mmHg (no dialysis hypotensive complication events are found based on the current defined standard), the loosest standard for the 100 - 110 mmHg range is that the pre-dialysis systolic blood pressure drops to 100 mmHg (no dialysis hypotensive complication events are found based on the current defined standard). The calculation formula for the initial decrease ratio value corresponding to the corresponding range is as follows:

[0097] α imin = 10 / ξ imax (ξ imax <= 110 mmHg)

[0098] From the above principle, it can be seen that the initial value of the blood pressure decrease ratio corresponding to each pre-dialysis systolic blood pressure value range is a relatively loose standard for the definition of hypotension. Then, the strictest standard blood pressure decrease ratio value, which is also the maximum ratio value α imax is:

[0099] α imax = (ξ imax - 90) / ξ imax

[0100] where ξ imax is the maximum value of the current dialysis systolic blood pressure value range, and α imax is the maximum decrease ratio value for each range.

[0101] When analyzing the pre-dialysis systolic blood pressure range greater than 180 mmHg, the initial value and the maximum value of α refer to the ratio value corresponding to the range where the pre-dialysis systolic blood pressure value is 180 mmHg. Therefore, according to the above principle, the corresponding blood pressure decrease ratio value α for the pre-dialysis systolic blood pressure value range is shown in Table 2.

[0102] Table 2

[0103]

[0104] Test analysis

[0105] In this embodiment, 6742 sample case data are used as fitting analysis data, and 2359 cases are recorded as verification data. Using different blood pressure drop ratio values α, data record distributions of different types are generated. After the fitting results of the algorithm model and the comprehensive analysis by combining the corresponding complication reports for verification and checking, the blood pressure drop ratio values α, accuracy rates, and AUCs corresponding to each pre-dialysis systolic blood pressure interval are shown in Table 3 as follows.

[0106] Table 3:

[0107] Pre-dialysis systolic blood pressure range Optimal blood pressure reduction ratio Accuracy of validation set AUC of validation set 90 - 100 mmHg 0.13 96.20％ 0.82 100 - 110 mmHg 0.14 95.29％ 0.81 110 - 140 mmHg 0.20 98.54％ 0.91 140 - 160 mmHg 0.26 96.19％ 0.85 160 - 180 mmHg 0.32 99.11％ 0.95 >180 mmHg 0.34 90.91％ 0.82

[0108] Result analysis:

[0109] As can be seen from the above results, currently, by using the algorithm model to mine the definition standard of dialysis hypotension that conforms to the current patient group, the results are relatively ideal, with the AUCs all greater than 0.8 and the verification accuracy rates greater than 90%. By using the method of this embodiment to mine and update the optimal definition of dialysis hypotension, and then predicting hypotension during hemodialysis for patients, the accuracy of the prediction results is improved.

[0110] In summary, a method for determining the definition value of hypotension during hemodialysis in this embodiment can obtain the definition value of dialysis hypotension suitable for the current patient group, solve the splitting of the discrimination of the same dialysis record by multiple definitions, and the new definition and analysis strategy can make the discrimination of dialysis hypotension cases have standard unity, improve the accuracy of predicting the occurrence probability of hypotension during dialysis, provide a basic reference for studying the causes of hypotension and formulating corresponding preventive measures, help prevent the occurrence of dialysis hypotension clinically, and have very important significance for improving the dialysis quality of hemodialysis patients and ensuring the safety of dialysis patients during the dialysis process.

[0111] Embodiment 2

[0112] This application embodiment provides a computer device, which includes a processor and a memory. At least one instruction or at least one program segment is stored in the memory, and the at least one instruction or the at least one program segment is loaded and executed by the processor to implement a method for determining the definition value of hypotension during hemodialysis as provided in the above method embodiment.

[0113] Figure 3 The following shows a schematic hardware structure diagram of a device for implementing a method for determining the definition value of hypotension during hemodialysis provided in this application embodiment. The device can participate in forming or include the device or system provided in this application embodiment. As Figure 3As shown, the computer device 10 may include one or more processors 1002 (the processors may include, but are not limited to, processing devices such as microprocessor MCUs or programmable logic devices FPGAs), a memory 1004 for storing data, and a transmission device 1006 for communication functions. In addition, it may further include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, a power supply, and / or a camera. Those of ordinary skill in the art can understand that Figure 3 the structure shown is only schematic and does not limit the structure of the above-mentioned electronic device. For example, the computer device 10 may further include more or fewer components than Figure 3 shown therein, or have a different configuration from Figure 3 that shown.

[0114] It should be noted that the above one or more processors and / or other data processing circuits are generally referred to as "data processing circuits" herein. The data processing circuit may be embodied in whole or in part as software, hardware, firmware, or any combination thereof. In addition, the data processing circuit may be a single independent processing module, or be incorporated in whole or in part into any one of other elements in the computer device 10 (or mobile device). As involved in the embodiments of the present application, the data processing circuit is a kind of processor control (such as the selection of a variable resistance terminal path connected to an interface).

[0115] The memory 1004 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to a method for determining the hypotensive definition value during hemodialysis in the embodiments of the present application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 1004, that is, implements the above-mentioned method. The memory 1004 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memories. In some instances, the memory 1004 may further include a memory remotely set relative to the processor, and these remote memories can be connected to the computer device 10 through a network. Examples of the above-mentioned network include, but are not limited to, the Internet, enterprise intranets, local area networks, mobile communication networks, and combinations thereof.

[0116] The transmission device 1006 is used to receive or send data via a network. Specific examples of the above-mentioned network may include a wireless network provided by a communication provider of the computer device 10. In one example, the transmission device 1006 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 1006 can be a Radio Frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0117] The display can be, for example, a touch-screen liquid crystal display (LCD), which enables a user to interact with the user interface of the computer device 10 (or mobile device).

[0118] Embodiment 3

[0119] The embodiment of the present application also provides a computer-readable storage medium, which can be disposed in a server to store at least one instruction or at least one segment of a program related to a method for determining a hypotensive definition value in hemodialysis in the method embodiment. The at least one instruction or the at least one segment of the program is loaded and executed by the processor to implement the method for determining a hypotensive definition value in hemodialysis provided in the above method embodiment.

[0120] Optionally, in this embodiment, the above storage medium can be located in at least one of multiple network servers in a computer network. Optionally, in this embodiment, the above storage medium may include, but is not limited to: various media that can store program codes such as a USB flash drive, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk, or an optical disc.

[0121] Embodiment 4

[0122] The embodiment of the present invention also provides a computer program product or a computer program. The computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the method for determining a hypotensive definition value in hemodialysis provided in the above various optional embodiments.

[0123] It should be noted that: the above order of the embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. And the above specific embodiments of the present application have been described. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require the particular order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0124] Each embodiment in the present application is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. The key point of each embodiment is to illustrate the differences from other embodiments. In particular, for the embodiments of the device, equipment, and storage medium, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiments.

[0125] Those of ordinary skill in the art can understand that all or part of the steps to implement the above embodiments can be completed by hardware or by a program instructing the relevant hardware. The said program can be stored in a computer-readable storage medium, and the above-mentioned storage medium can be a read-only memory, a magnetic disk, an optical disk, etc.

[0126] Taking the above ideal embodiments based on the present invention as inspiration, through the above description, relevant staff can completely make various changes and modifications without departing from the technical idea of this invention. The technical scope of this invention is not limited to the content in the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A method for determining the definition value of hypotension during hemodialysis, characterized in that, The method includes the following steps: S1. Obtain sample case data, and preprocess the sample case data to construct a data set; S2. Statistically analyze the data set, and based on the definition of dialysis hypotension and the data analysis strategy, combined with expert opinions, obtain the associated index factors for the definition of hypotension; S3. Input the data corresponding to the associated index factors for the definition of hypotension into the AI prediction model for processing and analysis to obtain the AUC value and the accuracy rate; S4. Determine the set of dialysis hypotension definition values based on the AUC value and the accuracy rate; S5. Verify the set of dialysis hypotension definition values with the reports on dialysis hypotension complications recorded by medical staff to obtain a subset of the set of dialysis hypotension definition values, so that when each definition value in the current subset is used as the dialysis hypotension standard recorded in the verification set data, it is determined whether the set of dialysis hypotension events in the data set completely includes the in - dialysis hypotension events recorded by medical staff; If it includes, use the maximum value in the subset of the set of dialysis hypotension definition values as the standard definition value for dialysis hypotension; If it does not include, find a maximum dialysis hypotension definition value from the set of dialysis hypotension definition values such that all values in the set of blood pressure reduction values for hypotension records in the complication report are greater than this dialysis hypotension definition value, and this dialysis hypotension definition value is the standard definition value for hypotension in the current pre - dialysis systolic blood pressure range.

2. The method for determining the definition value of hypotension during hemodialysis according to claim 1, wherein, The step S1 includes: Collect the pre - dialysis and in - dialysis record data from the LIS system, HIS system, and dialysis management system to form the sample case data, and the sample case data includes the patient's dialysis basic data and test data; The preprocessing of the sample case data includes: Collect and clean the patient's dialysis basic data and test data based on time logic to form a data set within an effective time period.

3. The method for determining the hypotensive definition value during hemodialysis according to claim 1, wherein, The step S2 includes the following steps: S21. Set the blood pressure drop ratio value α, and divide the data in the data set into positive and negative cases of hypotension according to the definition of dialysis hypotension corresponding to the blood pressure drop ratio value α. After division, hypotensive dialysis data and non - hypotensive dialysis data are formed; S22. Optimize the hypotensive dialysis data and non - hypotensive dialysis data; S23. Through statistical analysis of the distribution laws of the optimized hypotensive dialysis data and non - hypotensive dialysis data, combined with expert opinions, obtain the characteristic index factor variables related to whether dialysis hypotension occurs; S24. Screen the characteristic index factor variables to obtain the associated index factors for the definition of hypotension.

4. The method for determining the hypotensive definition value during hemodialysis according to claim 3, wherein The step S3 specifically includes the following steps: S31. Obtain the optimal blood pressure drop ratio value α from the dialysis systolic blood pressure value range [α imin , α imax ; ir ; S32. Input the data corresponding to the associated index factors for the definition of hypotension into the AI prediction model; S33, the AI prediction model re - fits and validates the analysis of the data corresponding to the associated index factors defined for hypotension based on the optimal blood pressure drop ratio value α ir and the data analysis strategy, and obtains the AUC value and accuracy rate.

5. The method for determining the definition value of hypotension during hemodialysis according to claim 4, wherein The step S4 specifically includes the following steps: If the AUC value is less than or equal to the preset value, adjust the dialysis hypotension definition value and return to step S2; If the AUC value is greater than the preset value, save the current hemodialysis hypotension definition value, update the data set or optimize the AI prediction model, and repeat steps S1 to S4 until the AUC value is greater than or equal to the expected value and the blood pressure drop ratio value α reaches the strictest standard definition value. Then, stop optimizing the data set or the AI prediction model, obtain the AUC values and accuracies corresponding to all saved hemodialysis hypotension definition values, compare them with the preset conditions to obtain a comparison result, and form a set of hemodialysis hypotension definition values based on the comparison result.

6. The method for determining the hypotensive definition value during hemodialysis according to claim 1, wherein In the step S31, α imin is calculated by the following formula: α imin = 20 / ξ imax (ξ imax (> 110 mmHg); α imin = 10 / ξ imax (ξ imax ≤ 110 mmHg); α imax The calculation formula is as follows: α imax = (ξ imax - 90) / ξ imax ; Among them, ξ imax is the maximum value of the current dialysis systolic blood pressure value range, α imin is the initial decrease ratio value of each interval, α imax is the maximum decrease ratio value of each interval.

7. The method for determining the definition value of hypotension during hemodialysis according to claim 1, characterized in that, The construction of the AI prediction model includes the following steps: Construct a data model and update the hypotension label value in the data record set according to the current definition of hemodialysis hypotension. Use the algorithm model to fit the data corresponding to the associated index factors of the hypotension definition, and use the particle swarm optimization algorithm to optimize the parameters of each algorithm model. Compare the prediction results of multiple locally optimal algorithm models and select the locally optimal algorithm model as the mining model. Input data into the mining model for inference, check and calculate and screen the inference results and the hypotension complication records in the dialysis record report to obtain the optimal inference result and the optimal hemodialysis systolic blood pressure drop ratio value. Verify the mining model based on the inference result. If the inference result is satisfied, the mining model is the AI prediction model.

8. The method for determining the definition value of hypotension during hemodialysis according to claim 3, wherein In step S22, principal component analysis is used to optimize the hypotensive dialysis data and non-hypotensive dialysis data through repeated dimensionality reduction and dimensionality increase operations.

9. A computer device, characterized in that, It includes: A processor; A memory for storing executable instructions; Among them, the processor is used to read the executable instructions from the memory and execute the executable instructions to implement the method for determining the hemodialysis hypotension definition value according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the processor is enabled to implement the method for determining the hemodialysis hypotension definition value according to any one of claims 1 to 8.

Images